About 40-60% of methadone maintenance patients are also cocaine dependent. Cocaine dependence is associated with significant morbidity and mortality, but few traditional therapies are efficacious in treating cocaine dependence in this difficult patient population. Contingency management (CM) strategies that provide positive incentives upon direct evidence of cocaine abstinence are promising interventions. Typically, vouchers, exchangeable for retail goods and services, are used as reinforcers. When voucher amounts range from $1000 to $3000 over a 12-week treatment period, CM can reduce cocaine use in methadone patients. We have data from cocaine-dependent patients treated in drug-free settings that suggest a novel reinforcement system that provides the chance to win prizes, rather than vouchers, may also be efficacious in decreasing cocaine use, at potentially lower costs. The purpose of the proposed study is to evaluate the efficacy of voucher and prize CM in cocaine-dependent methadone patients. Cocaine-dependent methadone patients (n=240) will be randomly assigned to one of four conditions: standard treatment, standard treatment plus usual magnitude prize CM ($300), standard treatment plus enhanced prize CM ($900), or standard treatment plus voucher CM ($900). Urine samples will be screened 2-3 times weekly for 14 weeks, and follow-up data will be collected throughout a 12-month period. We expect that CM will decrease cocaine use relative to standard treatment, the efficacy of prize CM will be magnitude dependent, and $900 prize CM will be more efficacious than $900 voucher CM. We will also examine patient characteristics and their association with treatment response. We expect that patients who submit one or more cocaine-negative samples during a pretreatment baseline phase will show good response to all CM conditions, but patients who submit all cocaine-positive samples during the baseline phase will benefit most from the high magnitude prize CM condition. Further, we will obtain a detailed analysis of relapse following CM treatment and evaluate the cost-effectiveness of CM. In sum, this study will provide a stringent test of the relative efficacy and cost-effectiveness of voucher and prize CM, and it will address moderators of response to CM in the treatment of cocaine-dependent methadone patients.